> Markdown version of [/jobs/ext/4746973-ml-systems-engineer-large-scale-model-training-rl-infrastructure](https://www.wearedevelopers.com/jobs/ext/4746973-ml-systems-engineer-large-scale-model-training-rl-infrastructure). Every page supports `.md` or `Accept: text/markdown`. Links point to the HTML versions so they work for humans too. Agent guide: [/agents.md](https://www.wearedevelopers.com/agents.md). --- # ML Systems Engineer, Large-Scale Model Training & RL Infrastructure - **Company:** Nebius Group - **Location:** Palo Alto, CA, United States (Remote available) - **Experience:** Expert - **Salary:** $195,200 - $262,200 - **Contract:** Permanent contract - **Skills:** Artificial Intelligence, Systems Engineering, Cloud Computing, Nvidia CUDA, Software Debugging, Software Design Documents, Distributed Systems, Memory Management, InfiniBand, Python (Programming Language), Remote Direct Memory Access, Graphics Processing Unit (GPU), Pytorch, DeepSpeed, Kubernetes, Free and Open-Source Software, Slurm, Machine Learning Operations, Megatron - **Published:** November 16, 2026 - **Apply:** https://www.careerbuilder.com/job-details/ml-systems-engineer-large-scale-model-training-rl-infrastructure-palo-alto-ca--9ae64910-67f1-4f36-b43e-9dcd8eac82fc ## About the Role Systems Engineering, Large-Scale Systems, Artificial Intelligence (AI), Cloud Computing, Data Modeling, GPU (Graphics Processing Unit), Research & Development (R&D), Hubs, Distributed Computing, Performance Modeling, * Strong Python and PyTorch engineering skills. * Hands-on experience with distributed model training, large-scale ML systems, or GPU cluster workloads. * Practical understanding of transformer training bottlenecks, memory pressure, gradient/optimizer state, communication overhead, and checkpointing. * Experience debugging production or research training jobs across multiple GPUs or nodes. * Ability to reason quantitatively about throughput, utilization, memory, reliability, cost, and research velocity. * Strong communication skills and ability to collaborate with researchers, ML engineers, platform engineers, and leadership. Nice-to-haves: * Experience with Megatron-LM, DeepSpeed, PyTorch FSDP/DTensor, Ray, Slurm, Kubernetes, or large internal training platforms. * Experience with RL infrastructure frameworks such as verl, slime, AReaL, OpenRLHF, TRL, or custom PPO/GRPO/RLHF systems. * Familiarity with NCCL, CUDA, Triton, Nsight, InfiniBand, RDMA, RoCE, H100/H200/B200 clusters, or storage/network bottlenecks. * Experience supporting SFT, DPO, PPO, GRPO, RLAIF, reward model serving, rollout generation, or agent training workloads. * Open-source contributions to distributed training, RL infrastructure, PyTorch, Ray, Megatron, DeepSpeed, or related systems., Applicants must be authorized to work in the country in which they apply and will be required to provide proof of employment eligibility as a condition of hire. ## Description Nebius Token Factory is building an AI training and model post-training capability for frontier model improvement. This role owns the infrastructure that makes large-scale training and RL experiments possible, reliable, reproducible, and efficient. The work sits at the intersection of distributed systems, GPU performance, model training frameworks, RL pipelines, and production engineering. A Senior ML Systems Engineer owns substantial training or RL infrastructure components end to end. They are deeply hands-on, can debug difficult distributed training failures independently, and can deliver measurable improvements in experiment throughput, stability, and GPU utilization. Your responsibilities: * Build and maintain distributed training infrastructure for SFT, continued pretraining, preference optimization, and RL workloads. * Integrate and extend frameworks such as Megatron-LM, DeepSpeed, PyTorch FSDP/DTensor, Ray, verl, slime, AReaL, OpenRLHF, or equivalent internal systems. * Implement and debug parallelism strategies including tensor, pipeline, sequence/context, expert, and data parallelism. * Build reliable rollout, reward model serving, replay/data buffer, checkpointing, evaluation, and experiment orchestration components for RL training. * Profile and improve GPU utilization, communication efficiency, memory usage, and training throughput. * Diagnose failures across NCCL, CUDA, PyTorch, Ray, schedulers, storage, networking, and checkpointing layers. * Create reproducible training runs, launch scripts, dashboards, runbooks, and operational tooling for research users. * Partner with research scientists to turn algorithmic training recipes into scalable, debuggable systems. * Write clear design docs, incident reports, benchmark reports, and operating guides.